Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/maxbarsukov/vkr-builderWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/maxbarsukov/vkr-builder/vkr-authoring)<a href="https://agentmods.dev/rules/maxbarsukov/vkr-builder/vkr-authoring"><img src="https://agentmods.dev/badge/rules/maxbarsukov/vkr-builder/vkr-authoring/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/maxbarsukov/vkr-builder/vkr-authoring"><img src="https://agentmods.dev/badge/rules/maxbarsukov/vkr-builder/vkr-authoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00766 |
| Opus 5.5 | $0.00000 | $0.00306 |
| Sonnet 5.5 | $0.00000 | $0.00153 |
| Haiku 4.5 | $0.00000 | $0.00077 |
Grade A, and why
vkr-authoring scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing thesis Markdown
Full specification: docs/llm-format/README.md (ru) / docs/llm-format/README.en.md (en).
Output Markdown only; the sole HTML allowed is the <!-- @suppress -->
directive.
Structure
- One logical section per file under the profile's
markdown_dir; the build order ismarkdown_filesinconfig.yaml. A new file is invisible until listed there. _bundle.mdis generated bymergeand deleted afterwards — never edit, reference or commit it.- Headings:
# 1 Название,## 1.1 Подраздел, no period after the number. Structural sections keep their canonical Russian caps names (# ВВЕДЕНИЕ,# СПИСОК ИСПОЛЬЗОВАННЫХ ИСТОЧНИКОВ,# ПРИЛОЖЕНИЕ А). Appendix letters skip I, O and Ё З Й О Ч Ь Ъ.
Numbered objects always use a key

Рисунок {pipeline} - Конвейер обработки
Таблица {req} - Требования
| Параметр | Значение |
|---|---|
| A | B |
- Figure caption immediately after the image, no blank line between them.
- Table and listing captions immediately before their block.
- Spaced separator
-or–after the key; keys short, lowercase, unique per kind. Numbers are assigned at build time. - A listing may be pulled from a file instead of pasted:
@listing md.pyon its own line under the caption,@listing md.py:582-607for a fragment. Path is relative tolistings_dir(orimages_dirwhen unset).
References, citations, formulas
-
Always by key:
[рис:pipeline],[табл:req],[лист:split],[форм:count]. Never hard-code a number. -
Citations
[1],[1; 2],[1-3];[1, 2]is not a GOST citation. Keyed style ([{gruber}]) is an alternative — do not mix the two. -
Escape literal brackets in prose:
\[note\],\{key\}. -
Display formula numbered by a key line directly under it:
\[ N = \sum_{i=1}^{k} 1 \] {count}
Tables and lists
Pipe tables only; every body row must match the header's cell count (that is
an error, not a warning). - , 1), а) for lists, two-space indent for
nesting.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 79 lines · 0 tokens per session scan A b708fa135535
vkr-authoring is a cursor rule published in the GitHub repository maxbarsukov/vkr-builder (9 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 766 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
vkr-authoring compared with cursor rules from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| vkr-authoringmaxbarsukov/vkr-builder | Cursor rule | 9 | 1mo ago | MIT | A | 0 | Cursor |
| mempalace-recall-alwaysMemPalace/mempalace | Cursor rule | 60k | today | MIT | A | 260 | Cursor |
| pythonPatrickJS/awesome-cursorrules | Cursor rule | 41k | 5mo ago | CC0-1.0 | A | 734 | Cursor |
| djangosanjeed5/awesome-cursor-rules-mdc | Cursor rule | 3.6k | 5mo ago | CC0-1.0 | A | 2,043 | Cursor |
| pythonRenvia-code/best-cursor-rules | Cursor rule | 13 | 11mo ago | CC0-1.0 | A | 0 | Cursor |
| api-testerTreevu-ai/cli-market-world | Cursor rule | 1 | 19d ago | none | A | 23 | Cursor |
| coding_styleEdison-Watch/open-edison | Cursor rule | 290 | 9mo ago | GPL-3.0 | A | 0 | Cursor |